Efficient production of graphene is one of the hurdles manufacturers are actively attempting to overcome. Typically, the quality and properties of graphene produced for use in a product remain unknown until after it’s made.
A team of researchers at Monash University have pioneered a machine learning-based method that can identify graphene properties and quality within just 14 minutes.
The team developed an algorithm using the data set of an optical microscope, which when turned loose on graphene, is capable of characterizing the material. The findings could help manufacturers of graphene or graphene oxide improve their processes and get a better handle on the quality of their product during production, including properties like thickness and atomic layer size.
“Graphene possesses extraordinary capacity for electric and thermal conductivity. It is widely used in the production of membranes for water purification, energy storage and in smart technology, such as weight loading sensors on traffic bridges,” said Professor Mainak Majumder, of Monash University’s Department of Mechanical and Aerospace Engineering.
One reason graphene is so expensive to produce at large quantities is the time-consuming process to ensure quality and uniformity across a batch. The research team’s findings could eliminate that hurdle and pave the way to faster, more efficient production of graphene.